About Pulsewidth, sync and waveforms. Issues and suggestions.

[LBH]

See this Driver image. Notice Pulsewidth = 10 - 90.: http://modularsynthesis.com/moog/921a/921a_oscillator_driver.jpg

Pulsewidth is originaly used for Square waves to make pulse waves.
It's logical that the same kind of pulsewidth do not apply to other waveforms. But some kind of morphing or other things can find place using Pulsewidth knob and PWM. It could also be possible to find a effect for the Sine wave.

Pulsewidth should not be zero or 100% as the sound is bad when to high and to low as the waveform close/ vanish. But perhaps between 2 and 98 as MOOG apparantly used pulse waves over 50%. Instead of for instance 15 they used 85%. This should be applied in both Modular and MINI V as it make a sound difference just like it make adifference if a saw is ramp up or down even if it should'nt. It just do. So a way to reverse the Waveforms can be useful too.
Also take a look at my last post with some links in this thread: http://forum.arturia.com/index.php?topic=87650.0.

The way pulsewidth is made for SAW wave in Modular V3, then it shift octave on 49 on the current scale from 0-50%. Try using a single osc to verify this. And also use a Tuner.
Pulsewidth should not affect the pitch to change a octave or any other musical interval. So if the current SAW pulsewidth that actually works like a sync effect shall remain (I think many like it), then it should not change octave at all if a 2-98% scale is used.

-----

In this video hardsync can be seen on a osciloscope. There are examples for Sine, Triangle and Saw.
Hardsync Video. Scope.: https://www.youtube.com/watch?v=z1cdVGKbzdw
It's for instance important to distinguish between Sync, Detune and FM.

For Sine Waves there is none pulsewidth effect of any kind in Modular V.
But worse the Sine wave don't have a Sync effect when synced. It just Detune. There should be a sync effect. It's possible as the video show. (Link above for Hardsync Video.)

Sine waves can not be hardsync to any other waves in Modular. All other waves i for now think work using Sync.


Best

[corum_beasley]

Until i read this post i had not tried the hard / soft sync button.

I watched the hard sync video https://www.youtube.com/watch?v=z1cdVGKbzdw to find out about it. I was a little confused at first but to help me understand a little better i set up the following on the modular v, see attached photo. I started setting the reference sync oscillator in this case 1, with zeroed controls and PWM to maximum, i set all the oscillators to plus 2 semitones and synchronized them to oscillator 1.

They behaved like this oscillator 2 exhibited hard / soft when the switch was flipped all other oscillators 3 to 9 did not.

I then one by one set each of the other oscillators as a reference in turn the behaviour was the same. It is always the one to the right and none of the others.

Like this ref 1 synchronized with 2, 2 with 3, 3 with 4, 4 with 5, 5 with 6, 6 with 7, 7 with 8, 8 with 9 and 9 with 1

I dont understand this behaviour, i chose the PWM setting and the tuning difference for convenient setup but i tried other values of PWM and slight detunes using the fine control, the behaviour was the same.

Surely any of the other oscillators can be hard / soft synchronized with any of the others or do i not understand how this works?

[LBH]

You can only sync one oscilator with one other oscilator. for instance sync osc 2 to osc 1 - or - osc 8 to osc 1 etc...
If you try to sync more osc's to a single master like for example trying to sync osc 2 and 8 to osc 1, then only osc 2 is synced.

There seem to be a "The osc nearest to the right" priority for sync if mulitiple osc's is tried to be synced to the same osc. (Osc 1 is right of osc 9.)

I do believe it's normal having one master and one slave osc only when syncing. If it's intended to have multiple Slave osc's synced to a single Master osc to be possible in Modular - like the possibility to select a Master osc several times in the menu indicate - , then this do not work. I would think it is not intended to work, so it should not be possible to select a osc for Master more than once. The osc selection should be toned out once it's selected or at least marked to show it has already been used for Master in the Sync menu.

It's possible to sync in series like osc 2 to osc 1 and at the same time sync osc 3 to osc 2. But i don't think this is a "correct" posibility, and not sure if the sound results will not end in unwanted noise. However the different possible connections might also create interesting results at times.
If this shall not be possible, then any osc that is in a sync connection should be toned out or at least marked to show it has been used for Master and/ or Slave in the sync menu, so those osc's can't be selected and used twice - or just to let you know that the Osc have been used in a sync connection once and how it is used, to be warning the user about unspected behaviors can occour.

It's also possible - and more "correct" i think - , that you for instance sync osc 8 to osc 2 and osc 9 to osc 1 etc. at the same time. That mean - connections that don't involve a single osc more than once and only either as a Master or a Slave osc. Having 9 osc's then this can be done 4 times using 4 pairs of osc's (if you for any reason shuould want to do that), and having a osc left.

At least a marking of some kind in the sync menu if osc's are used as Masters and/ or Slaves should be done.
If it should not be possible to use a osc more than once either it's as Master or Slave in a sync connection, then this type of connections should be made impossible to do. But if it is intended to be possible, then there is something wrong.

If anyone have other knowledge about this, then please join in.


Best

[jeffbart]

Re the pulse width of sawtooth
I think the method Arturia may use to change the pulse with on square waves may work differently on the sawtooth wave . It appears/sounds like the end of the cycle and start point of next cycle is e.g. moved more and more to the left so that in the space of the one large cycle of sawtooth you get two smaller instances of a sawtooth cycles, and at some point there are two identical half sized sawtooth cycles in the space of one.....which sounds an octave higher. I don't find this any kind of bug, I think it's an added 'feature'. "Is not bug, is feature".
cheers jb

[LBH]

Where have anyone called the saw pulsewidth whole range behavior/feature a bug?
I think there is a misunderstanding of what is said in the posts and about what the points are.
What i say about the octave shift is it introduce some odd things, that i even suggested how it's perhaps is possible to go around/ fix.
And then the thread is about others things too. But it's not about removing the saw pw feature if it can be kept without problems, like i suggest how too.

Please read the posts again.

Also i can suggest to read the manual at the bottom of page 92 - starting like this: "The synchronization of an oscillator on another brings about complex waveforms........" and the top of page 93 - ending like this: "In the image above, oscillator2 is synchronized with the first, and then tuned to a frequency with double the tonality."
Do this described sync match the readings using pulsewidth for a saw wave? To me it do. That's why i call the PW feature for the saws a sync effect. And you can create a Sync modulation effect for a saw using the PW knob.
Keep in mind that all pw effects for other waveforms than square/ pulse waves are features.

The values where the saw not just sound like it's an octave hignher, but actually is higher when using a tuner is around the value 0.48 - 0.49 (perhaps earlier) on the current PW knobs scale as i have described in my first post. That's nearly fully to the right. Use a single osc using a saw to verify this. Use a tuner and see it change octave when nearly at fully right.
Please also keep in mind that the length of a waveform as seen in a scope is related to pitch.

A problem with the octave shift can occour like this. You wan't a full normal saw in osc 1 and a full square in osc 2 and you wan't them to sound on the same octave. To get a full saw you need to have the saw fully left to sound in the octave set by the range knob for instance 8'. To get a full square you need to have the pulsewidth set to fully right, and at that point it sound like the range you have set - like for instance 8'. But if you have the pulsewidth fully right to get that square, then at that setting the saw is actually a octave higher. I not just sound like it's higher. it is a octave higher as a tuner also can show. So to get the saw to be the same octave as a 8' square, then you have to set the saw range to 16' to get the same octave for both waveforms.
If you set the pw fully left to match get a clean saw that match the set range like for instance 8', then you can only use saw and sine in that driver section for anything, as both square and triangle will give no sound in that PW position.

This octave change also introduce issues in other relations, like when syncing osc's, if you are not aware of this octave change, as it do matter if for instance a slave osc in any sync relation is tuned a octave lower than the master.
Only when syncing two saws in the same driver section may not cause issues regarding this. But it can still matter in other ways.

Regarding the PW of the saw (or for any other waveforms effect/ feature) - i just suggest the effect/ feature applied -  should be adapted and have set limits so unwanted behaviors caused by the feature is ruled out, if the effect/ feature shall remain.

Also remember my posts is not just about the above. That's just a parts of it.
It's also about PW can be set to 0 where there is no sound for triangle and square waves. (But that's where the saw is correct as a clean saw tuned in the correct octave. There need to be made some corrections about some things here - also to keep the saw pw feature. More in my previous posts.)
And it's about why Sine waves have no PW feature at all, and very much about why Sine waves can't be used in sync pairing if it actually is possible.

NB: Another way to fix the PW at zero issue is if the SAW was clean and tuned correct in the PW knobs fully right position, and then not allow values under 0.03 or 0.02 for pulsewidths.
Please also read my suggestions and comments for PW values in the first post of this thread.


Cheers
LBH